1. Field of the Invention
The present invention relates to a color-adjusting apparatus that adjusts color data for a color reproduction device, such as a TV apparatus, a mobile terminal, a computer display, and a color printer, and arts related thereto. More particularly, the present invention relates to a technique that adjusts a specific color and a total color of an inputted image.
In this specification, all kinds of physical quantity indicating how a color is bright, such as lightness, value, luminance, and so on, is generically called “brightness”. Similarly, all kinds of physical quantity indicating how a color looks vivid, such as chromaticity, saturation, and so on, is generically called “chroma”. Note that a color is a three-dimensional vector in a color space, basically.
2. Description of the Related Art
For example, a user of the TV apparatus can adjust a chroma phase and a chroma level of analog video signals reproduced thereby, utilizing a conventional color-adjusting apparatus mounted therein. Furthermore, the user can adjust a total color of the TV apparatus to his/her favorite color, changing a level of brightness of the analog video signals.
Adjusting the total color is effective, when difference of conditions, such as lighting environment, color characteristics of a camera and/or a display device, should be absorbed to perform transparent reproduction, and when chroma should be increased to reproduce colors more vividly. Once the total color has been adjusted suitably, the user may be almost satisfied with a result of the color reproduction. In this specification, a “total” color is a color that is an object of the total color adjustment.
Meanwhile, it is known that a “memory color”, such as skin color, sky blue, grass green, and so on, which is well retained in human memory, differs from a color of the real thing corresponding thereto. In the so-called “preferred color reproduction” (for example, in reproduction of the “memory color”), it is insufficient to perform the total color adjustment merely. Furthermore, it is necessary to be able to adjust a current color to the “memory color”, individually. In this specification, a “specific” color is a color that is an object of the specific color adjustment.
A document 1 (published Japanese Patent Application Laid-Open No. H05-300531) discloses a color-adjusting method that adjusts a specific color region individually, by designating the specific color.
A document 2 (published Japanese Patent Application Laid-Open No. H06-78320) discloses a color-adjusting apparatus. The color-adjusting apparatus sets a weighting coefficient, according to difference between an inputted chroma value and a pre-selected reference chroma value. The color-adjusting apparatus then adjusts a particular subject of a color in a chroma plane, whose coordinates are hue components and chroma components, based on the weighting coefficient.
Furthermore, a document 3 (published Japanese Patent Application Laid-Open No. H10-198795) discloses a color-adjusting apparatus. The color-adjusting apparatus of the document 3 calculates a color approximation degree hx that indicates an approximation degree of the current hue to a designated hue.
The color-adjusting apparatus of the document 3 adjust a color of (R, G, B) to output a color of (R′, G′, B′), according to the following formula.(R′, G′, B′)=(R, G, B)+hx*(a1, a2, a3),where a1, a2, and a3 are vector-adjusting coefficients for R, G and B, respectively.
Referring to the documents 1, 2 and 3, a color-adjusting apparatus that adjusts both of the specific color and the total color may be constructed as shown in FIG. 13.
In FIG. 13, a total color-adjusting unit 1 performs the total color adjustment of data of an inputted color, and outputs a result vector of the total color adjustment to a composing unit 6 and a specific color-weighting coefficient-calculating unit 3. Thanks to the total color adjustment, the difference of conditions, such as lighting environment, color characteristics of a camera and/or a display device, has been absorbed.
A specific color-adjusting stage 2, which is provided next to the total color-adjusting unit 1, comprises the following elements. The specific color-weighting coefficient-calculating unit 3 calculates a weighting coefficient k with respect to a specific color, such as skin color, and so on.
A specific color-adjusting vector-outputting unit 4 outputs, to a multiplying unit 5, data of an adjusting vector δ with respect to the specific color. The multiplying unit 5 multiplies the adjusting vector δ by a scalar of the weighting coefficient k to output data of a weighted vector to the composing unit 6.
The composing unit 6 inputs the data of the result vector of the total color adjustment from the total color-adjusting unit 1, and inputs the data of the weighted vector from the multiplying unit 5.
The composing unit 6 adds the result vector and the weighted vector, and outputs data of an added vector. In short, first, the total color adjustment is made; secondly, the specific color adjustment is made. That is, with respect to the specific color, adjustments are duplicated.
It is assumed that the data of the inputted color has chroma components, each of which has 8-bit data length (256 scales in total). Furthermore, it is assumed that the chroma components should be multiplied by a factor of “1.2”, generally, and further that a chroma component of the specific color should not be changed.
In this case, the total color-adjusting unit 1 multiplies the chroma components of the inputted color by the factor of “1.2”, and outputs data of a result vector. The specific color-adjusting vector-outputting unit 4 and/or the multiplying unit 5 multiplies, by a factor of “1/1.2”, the chroma component of the specific color, the chroma component being included in the data of the result vector. That is, with respect to the specific color, duplication of adjustments cannot be avoided.
In digital image processing, since each of coordinates of the inputted color is quantized, quantization errors may occur in some cases. For example, choose an original value of “99”, multiplied by a factor of “1.2”, and multiplied by a factor of “1/1.2”; the result is “98”, which does not equal the original value. Also, after the duplicated adjustment, in many cases, a quantization error may occur, thereby the chroma component of the specific color may be changed, contrary to expectations.
Furthermore, in the digital image processing, a clip of a value may occur in some cases. When data has 8-bit length, a value of the data cannot be 256 or more. For example, choose an original value of “250”, multiplied by a factor of “1.2”; the result is not “300” but “255”, because a clip of the value has occurred. In addition, the result of “255” multiplied by a factor of “1/1.2” is “212”, which does not equal the original value.
In a case where the color-adjusting apparatus should adjust both of the specific color and the total color, since the adjustments must be duplicated according to the prior arts, quality of an inputted image is deteriorated very easily, caused by the quantization error and/or the clip of a value, and so on.
Meanwhile, it is assumed that brightness and/or chroma with respect to the specific color should not be changed, and further that hue of the specific color should be changed to more preferable one. In this case, since the total color-adjusting unit 1 has adjusted the total color before adjustment of the specific color, whole conversion loses linearity, and handling thereof is very difficult.